Integrate Data Mining with Decision Support Systems to Support Engineering Decision Making

Traditional engineering decision support systems (EDSS) cannot be well applied in the practice of construction management due to their inherent limitations in knowledge acquisition. In this paper, a new approach named DMDS is proposed, which integrates data mining technologies with decision support systems. DMDS aims to automatically extract knowledge from a large amount of historic data generated in engineering practice and then reason with the extracted knowledge to support engineering decision makings in construction. A prototype system named DM-EDSS has also been designed and developed to demonstrate the feasibility and effectiveness of the DMDS model. The findings demonstrate that at least the following benefits can be achieved: (1) providing a new approach for EDSS to automatically acquire knowledge from a large mount of engineering data; (2) enabling insights to be gained on factors that have impacts on construction problems; and (3) providing a uniform decision support process for different decision-making problems. KEY WRODS integration model, data mining, decision support system, engineering decision making, reasoning mechanism INTRODUCTION Due to the complexity and uniqueness of construction projects, most engineering decisions are experience-based, and may be affected by many factors such as decision makers’ knowledge backgrounds, experience, emotion and characters, etc. In order to improve the efficiency and effectiveness of engineering decision makings in construction, many engineering decision support systems (EDSS) that try to reuse field experts’ knowledge and experience to solve new engineering problems have been designed and developed. However, only few of them have been widely employed in the engineering practice because of their inherent limitations in knowledge acquisition. In traditional EDSS, the knowledge (e.g., rules or models) needed by their reasoning mechanisms have to be manually summarized from 1 Ph.D. Candidate, Civil Engineering Department, Tsinghua University, Beijing, China. Phone +86-1062778987, FAX +86-10-62784975, [email protected] 2 Professor, Civil Engineering Department, Tsinghua University, Beijing, China. Phone +86-10-62784975, FAX +86-10-62784975, [email protected] 3 Professor, Building and Real Estate Department, the Hong Kong Polytechnic University, Hong Kong, China. Phone +852-27665817, FAX +852-27645131, [email protected] June 14-16, 2006 Montréal, Canada Joint International Conference on Computing and Decision Making in Civil and Building Engineering